JP2022139311A - Automatic driving system - Google Patents

Abstract

To suppress a sense of unfairness among users when setting a target route for an automatic driving vehicle that provides a mobility service which transports a passenger.SOLUTION: An automatic driving system is applied to an automatic driving vehicle that provide a mobility service to transport a passenger. The automatic driving system performs recognition processing to recognize a surrounding state of an automatic driving vehicle using a recognition sensor mounted on the automatic driving vehicle. Also, the automatic driving system performs route setting processing to set a route to a destination or a loop route as a target route for the automatic driving vehicle. The route setting processing includes: route candidate acquisition processing for acquiring a plurality of target route candidates; and route selection processing for selecting the target route from among the plurality of candidates according to recognition performance of the recognition processing. In the route selection processing, the automatic driving system selects a candidate with a shorter distance as the target route as the recognition performance declines.SELECTED DRAWING: Figure 7

Description

The present disclosure relates to technology for setting a target route for an autonomous vehicle that provides passenger transport services.

US Pat. No. 6,300,003 discloses a route planning method for autonomous vehicles. An autonomous vehicle may stall while driving. Insufficient batteries, low communication quality areas, construction sections, bad weather, and the like are examples of factors that cause an autonomous vehicle to stall (stall factors). If there is a stall factor on the route to the destination and there is no user of the autonomous vehicle on that route, an alternative route without the stall factor is set.

JP 2020-106525 A

Consider setting target routes for self-driving vehicles that provide mobility services to carry passengers. According to the technique disclosed in Patent Document 1, a target route is set so as to avoid stall factors. However, simply setting the target route while avoiding the stall factor may cause a sense of unfairness among users who are in similar road conditions.

For example, consider a case where the user wishes to ride at two locations with similar road conditions. If the target route is simply changed to avoid the stall factor, the autonomous vehicle may pass through one point but not the other. In that case, a feeling of unfairness arises between users who are in similar road conditions.

One object of the present disclosure is to provide a technology capable of suppressing a sense of unfairness among users when setting a target route for an autonomous vehicle that provides a mobility service for transporting passengers.

The first aspect relates to automated driving systems.
Self-driving systems are applied to self-driving vehicles that provide mobility services to carry passengers.
An automated driving system includes one or more processors.
The one or more processors are
Recognition processing that recognizes the surroundings of the autonomous vehicle using a recognition sensor mounted on the autonomous vehicle;
and a route setting process for setting a route to a destination or a loop route as a target route for the automated driving vehicle.
The route setting process is
route candidate acquisition processing for acquiring a plurality of target route candidates;
a route selection process for selecting a target route from among a plurality of candidates according to the recognition performance of the recognition process;
including.
In the route selection process, the one or more processors select shorter distance candidates as target routes as the recognition performance declines.

The second aspect relates to the target route setting method.
The target route setting method sets a target route for an autonomous vehicle that provides mobility services for carrying passengers.
The target route setting method is
Recognition processing that recognizes the surroundings of the autonomous vehicle using a recognition sensor mounted on the autonomous vehicle;
A route setting process for setting a route to a destination or a loop route as a target route for the automated driving vehicle.
The route setting process is
route candidate acquisition processing for acquiring a plurality of target route candidates;
a route selection process for selecting a target route from among a plurality of candidates according to the recognition performance of the recognition process;
including.
The route selection process selects candidates with shorter distances as target routes as the recognition performance declines.

According to the present disclosure, a target route is selected from multiple candidates according to the recognition performance of recognition processing. More specifically, target routes with shorter distances are selected as recognition performance decreases. This makes it easier to continue mobility services using autonomous vehicles even when recognition performance is degraded. That is, continuity of mobility services is ensured.

Moreover, according to the present disclosure, it is possible to suppress a sense of unfairness among users by considering the distance of the entire route. A target route with a short distance mainly passes through a main road and has a low probability of entering a community road. Therefore, users in similar road conditions will be provided with similar mobility services. Therefore, a sense of unfairness among users is suppressed.

1 is a conceptual diagram for explaining an outline of a mobility service according to an embodiment of the present disclosure; FIG. FIG. 2 is a conceptual diagram showing an example of route candidates for an automatic driving vehicle according to an embodiment of the present disclosure; FIG. 1 is a block diagram showing a configuration example of an automatic driving system according to an embodiment of the present disclosure; FIG. 4 is a block diagram showing an example of driving environment information in the embodiment of the present disclosure; FIG. It is a flow chart which shows processing related to automatic operation control by an automatic operation system concerning an embodiment of this indication. 4 is a flowchart showing route setting processing by the automatic driving system according to the embodiment of the present disclosure; 4 is a flowchart showing an example of route selection processing by the automatic driving system according to the embodiment of the present disclosure; 7 is a flowchart showing another example of route selection processing by the automatic driving system according to the embodiment of the present disclosure;

Embodiments of the present disclosure will be described with reference to the accompanying drawings.

1. Outline FIG. 1 is a conceptual diagram for explaining an outline of the mobility service according to the present embodiment. A mobility service is a service that transports passengers (users). In the present embodiment, the mobility service is provided by an automatically driving vehicle 1 capable of automatically driving (autonomous driving). For example, the self-driving vehicle 1 is an self-driving bus. The automatic driving here assumes that the driver does not necessarily have to concentrate on driving 100% (for example, so-called level 3 or higher automatic driving).

The automatically driven vehicle 1 travels to a destination or travels on a circuit route by automatic operation. Hereinafter, such an operation route on which the automatic driving vehicle 1 travels will be referred to as a "target route RT".

The automatic driving system 10 is a system applied to the automatic driving vehicle 1 and controls the automatic driving vehicle 1 . For example, the automatic driving system 10 is installed in the automatic driving vehicle 1 . Alternatively, at least part of the automated driving system 10 may be included in a remote system external to the automated driving vehicle 1 to remotely control the automated driving vehicle 1 . In other words, the automated driving system 10 may be distributed between the automated driving vehicle 1 and the remote system.

The automatic driving system 10 performs a “route setting process” for setting a target route RT for the automatic driving vehicle 1 . As will be described later, according to the present embodiment, the automatic driving system 10 can variably set the target route RT. Then, the automatic driving system 10 performs "automatic driving control" for controlling the automatically driving vehicle 1 to travel along the target route RT.

In automatic driving control, it is necessary to recognize the situation around the automatic driving vehicle 1 . The automatic driving vehicle 1 is equipped with a recognition sensor (external sensor) that recognizes surrounding conditions. Examples of the recognition sensor include a camera, LIDAR (Laser Imaging Detection and Ranging), radar, and the like. By using recognition sensors, the automatic driving system 10 detects pedestrians, other vehicles (preceding vehicles, parked vehicles, etc.), road structures (white lines, curbs, guardrails, walls, medians, roadside structures, etc.), recognize signs, etc. Then, the automatic driving system 10 performs automatic driving control based on the result of recognition processing using the recognition sensor 21 .

However, the recognition performance of recognition processing may deteriorate due to various factors. For example, recognition performance may degrade in weather conditions such as rain or snow. As another example, contamination of the recognition sensor may degrade recognition performance. As yet another example, recognition performance may be degraded due to recognition sensor malfunction.

If the recognition performance of the recognition process is degraded, the accuracy of automatic driving control based on the results of the recognition process may also be degraded. In consideration of this point, the automatic driving system 10 according to the present embodiment variably sets (switches) the target route RT of the automatically driving vehicle 1 according to the recognition performance.

FIG. 2 shows an example of candidates for the target route RT. A candidate for the target route RT is hereinafter referred to as a "route candidate". FIG. 2 shows three route candidates, a first route R1, a second route R2, and a third route R3. In the example shown in FIG. 2, the target route RT is a circular route, but the same applies to a route to a destination.

Among the three types of route candidates, the first route R1 is the longest and the third route R3 is the shortest. For example, the first route R1 not only passes through a relatively wide arterial road, but also enters a relatively narrow community road. On the other hand, the third route R3 mainly passes through relatively wide arterial roads and does not enter relatively narrow community roads. The second route R2 is intermediate between the first route R1 and the third route R3.

From the viewpoint of automatic driving control, the recognition performance required increases as the route distance increases. This is because as the route distance increases, the time and frequency of traveling on various types of roads including narrow community roads increases. Conversely, when the route distance is short, there is a high possibility that the automatic driving control can be continued even if the recognition performance is relatively low.

From the above point of view, according to the present embodiment, multiple route candidates are associated with recognition performance. The first route R1 with the longest route distance is associated with a state of high recognition performance. A second route R2 with an intermediate route distance is associated with an intermediate state of recognition performance. The third route R3, which has the shortest route distance, is associated with low recognition performance. It can be said that the first route R1 is the default target route RT, and the second route R2 and the third route R3 are alternative target routes RT when the recognition performance is degraded.

The automatic driving system 10 selects the target route RT according to the recognition performance of recognition processing. That is, the automatic driving system 10 selects the target route RT according to the recognition performance from multiple route candidates.

For example, when the recognition performance is high, the automatic driving system 10 selects the first route R1 with the longest route distance as the target route RT. When the recognition performance deteriorates, the automatic driving system 10 selects the second route R2 as the target route RT. When the recognition performance further deteriorates, the automatic driving system 10 selects the third route R3 as the target route RT. That is, the automatic driving system 10 selects a route candidate with a shorter distance as the target route RT as the recognition performance declines.

As described above, according to the present embodiment, the target route RT is selected from multiple route candidates according to the recognition performance of the recognition processing. For example, a target route RT with a shorter distance is selected as recognition performance declines. This makes it easier to continue the mobility service using the autonomous vehicle 1 even when the recognition performance is degraded. That is, continuity of mobility services is ensured.

Moreover, according to the present embodiment, it is possible to suppress a sense of unfairness among users by considering the distance of the entire route.

For example, consider a case where the user wishes to ride at two locations with similar road conditions. If the target route RT is simply changed to avoid obstacles (stall factors), there is a possibility that the autonomous vehicle 1 will pass through one point but will not pass through the other point. In that case, a feeling of unfairness arises between users who are in similar road conditions.

According to the present embodiment, a target route RT with a shorter distance is selected as recognition performance declines. A short target route RT (eg, the third route R3) mainly passes through main roads and has a low probability of entering community roads. Therefore, users in similar road conditions will be provided with similar mobility services. Therefore, a sense of unfairness among users is suppressed.

The automatic driving system 10 according to this embodiment will be described in more detail below.

2. Automatic driving system 2-1. Configuration Example FIG. 3 is a block diagram showing a configuration example of the automatic driving system 10 according to the present embodiment. The automatic driving system 10 includes a sensor group 20 , a travel device 30 and a control device 100 .

The sensor group 20 is mounted on the automatic driving vehicle 1 . The sensor group 20 includes a recognition sensor (external sensor) 21 that recognizes the surroundings of the autonomous vehicle 1 . Examples of the recognition sensor 21 include a camera, LIDAR (Laser Imaging Detection and Ranging), radar, and the like.

The sensor group 20 also includes a vehicle state sensor that detects the state of the automatically driven vehicle 1 . Vehicle state sensors include speed sensors, acceleration sensors, yaw rate sensors, steering angle sensors, and the like. Further, the sensor group 20 includes position sensors that detect the position and orientation of the autonomous vehicle 1 . A GPS (Global Positioning System) sensor is exemplified as a position sensor.

The traveling device 30 is mounted on the automatic driving vehicle 1 . The traveling device 30 includes a steering device, a driving device, and a braking device. The steering device steers the wheels. For example, the steering system includes a power steering (EPS: Electric Power Steering) system. A driving device is a power source that generates a driving force. An engine, an electric motor, an in-wheel motor, etc. are illustrated as a drive device. A braking device generates a braking force.

The control device 100 controls the automatically driven vehicle 1 . The control device 100 includes one or more processors 110 (hereinafter simply referred to as processors 110) and one or more storage devices 120 (hereinafter simply referred to as storage devices 120). Processor 110 executes various processes. For example, processor 110 includes a CPU (Central Processing Unit). The storage device 120 stores various information. Examples of the storage device 120 include volatile memory, nonvolatile memory, HDD (Hard Disk Drive), SSD (Solid State Drive), and the like. Various processes by processor 110 (control device 100) are realized by processor 110 executing a control program, which is a computer program. The control program is stored in the storage device 120 or recorded in a computer-readable recording medium. The control device 100 may include one or more ECUs (Electronic Control Units). A part of the control device 100 may be an information processing device outside the automatic driving vehicle 1 . In that case, part of the control device 100 communicates with the automatically driven vehicle 1 and remotely controls the automatically driven vehicle 1 .

2-2. Driving Environment Information The processor 110 acquires the driving environment information 200 indicating the driving environment of the automatically driving vehicle 1 using the sensor group 20 . Driving environment information 200 is stored in storage device 120 .

FIG. 4 is a block diagram showing an example of the driving environment information 200. As shown in FIG. The driving environment information 200 includes surrounding situation information 210, vehicle state information 220, map information 230, vehicle position information 240, weather information 250, and the like.

The surrounding situation information 210 is information indicating the surrounding situation of the automatically driven vehicle 1 . The processor 110 acquires surrounding situation information 210 using the recognition sensor 21 . Surrounding information 210 includes information obtained by the recognition sensor 21 . For example, surroundings information 210 includes image information captured by a camera. As another example, the surroundings information 210 includes point cloud information obtained by LIDAR.

Surroundings information 210 further includes object information 215 about objects around the autonomous vehicle 1 . Objects include pedestrians, bicycles, other vehicles (leading vehicles, parked vehicles, etc.), road structure (white lines, curbs, guardrails, walls, median strips, roadside structures, etc.), signs, poles, obstacles, etc. are exemplified. The object information 215 indicates the relative position and relative speed of the object with respect to the autonomous vehicle 1 . For example, by analyzing image information obtained by a camera, an object can be identified and the relative position of the object can be calculated. Also, based on the point cloud information obtained by LIDAR, an object can be identified and the relative position and relative velocity of the object can be obtained. The object information may include the moving direction and moving speed of the object.

The vehicle state information 220 is information indicating the state of the automatically driven vehicle 1 . Processor 110 obtains vehicle state information 220 from vehicle state sensors. The vehicle state information 220 may indicate the driving state (automatic driving/manual driving) of the automatically driven vehicle 1 .

Map information 230 includes a general navigation map. Map information 230 may indicate lane layout, road geometry, and the like. Map information 230 may include location information such as landmarks, traffic lights, signs, and the like. The processor 110 obtains the map information of the required area from the map database. The map database may be stored in a predetermined storage device mounted on the automatic driving vehicle 1, or may be stored in an external management server. In the latter case, processor 110 communicates with the management server to obtain the necessary map information.

The map information 230 may further include the target route RT of the autonomous vehicle 1 . The details of setting the target route RT will be described later.

The vehicle position information 240 is information indicating the position of the automatically driven vehicle 1 . Processor 110 acquires vehicle position information 240 from the detection result of the position sensor. Further, the processor 110 may acquire highly accurate vehicle position information 240 by well-known self-position estimation processing (localization) using the object information 215 and the map information 230 .

Weather information 250 indicates weather conditions such as rain, snow, fog, and the like. Weather information 250 may indicate rainfall, snowfall, fog density, and the like. For example, weather information 250 is provided by a weather information service system. Processor 110 obtains weather information 250 from the weather information service system via communication. As another example, processor 110 may obtain weather information 250 based on surroundings information 210 . For example, by analyzing image information captured by a camera, the weather can be determined and the weather information 250 can be obtained.

2-3. Vehicle Driving Control The processor 110 executes “vehicle driving control” for controlling driving of the automatically driven vehicle 1 . Vehicle running control includes steering control, acceleration control, and deceleration control. Processor 110 executes vehicle travel control by controlling travel device 30 (steering device, drive device, braking device). Specifically, processor 110 performs steering control by controlling a steering device. Processor 110 also performs acceleration control by controlling the drive. Processor 110 also performs deceleration control by controlling the braking device.

2-4. Processing Related to Automatic Driving Control The processor 110 controls automatic driving of the automatic driving vehicle 1 . FIG. 5 is a flowchart showing processing related to automatic driving control according to the present embodiment. The processing flow shown in FIG. 5 is repeatedly executed at fixed cycles.

In step S100, processor 110 performs an “information acquisition process” for acquiring driving environment information 200 described above. The information acquisition process also includes a "recognition process" for recognizing the surroundings of the automatic driving vehicle 1 (step S150). In the recognition process, the processor 110 uses the recognition sensor 21 to recognize the surrounding conditions of the automatically driven vehicle 1 and acquires the surrounding condition information 210 (especially the object information 215).

In step S<b>200 , the processor 110 performs automatic driving control based on the driving environment information 200 . At this time, the processor 110 performs vehicle travel control so that the automatically driven vehicle 1 travels along the target route RT.

More specifically, the processor 110 generates a driving plan for the automatically driving vehicle 1 based on the driving environment information 200 . The driving plan is exemplified by maintaining the current driving lane, changing lanes, turning left or right, avoiding obstacles, and the like. Furthermore, based on the driving environment information 200, the processor 110 generates a target trajectory (target trajectory) necessary for the autonomous vehicle 1 to travel according to the travel plan. A target trajectory includes a target position and a target velocity. Then, the processor 110 performs the vehicle travel control described above so that the automatically driven vehicle 1 follows the target trajectory.

3. Route setting process The processor 110 performs a “route setting process” for setting the target route RT of the autonomous vehicle 1 . The target route RT is a route to the destination or a circular route. The route setting process may be performed before the operation of the automatically driven vehicle 1 or may be performed while the automatically driven vehicle 1 is in operation. The route setting process may be performed at regular intervals.

FIG. 6 is a flowchart showing route setting processing by the automatic driving system 10 according to this embodiment.

3-1. Route Candidate Acquisition Processing (Step S310)
In step S310, processor 110 performs a "route candidate acquisition process" for acquiring a plurality of route candidates. Examples of multiple route candidates are as shown in FIG. Route candidate information 300 is information indicating a plurality of route candidates. The processor 110 acquires the route candidate information 300 and stores the acquired route candidate information 300 in the storage device 120 (see FIG. 3).

For example, multiple route candidates are predetermined by the mobility service administrator. In this case, the route candidate information 300 is generated by an administrator and provided to the automatic driving system 10 . The processor 110 acquires the route candidate information 300 from the administrator.

As another example, processor 110 may generate route candidate information 300 by itself. In this case, processor 110 searches for various route candidates based on map information 230 and obtains route candidate information 300 .

A plurality of route candidates may be pre-associated with recognition performance of the recognition process. In this case, the route candidate information 300 indicates route candidates and recognition performance in association with each other.

3-2. Recognition performance acquisition process (step S320)
In step S320, processor 110 performs a "recognition performance acquisition process" for acquiring the recognition performance of the recognition process (step S150). The recognition performance information 400 is information indicating recognition performance. The processor 110 acquires the recognition performance information 400 and stores the acquired recognition performance information 400 in the storage device 120 (see FIG. 3).

As a method of acquiring the recognition performance, various examples are conceivable as follows.

In a first example, processor 110 obtains recognition performance based on weather information 250 . For example, when it is sunny, the processor 110 determines that the recognition performance is high. If the stormy weather index (eg rainfall, snowfall, or fog density) exceeds the first threshold, processor 110 determines that the recognition performance is moderate. If the stormy weather index exceeds a second threshold that is greater than the first threshold, processor 110 determines that the recognition performance is poor. If the stormy weather index exceeds a third threshold that is greater than the second threshold, processor 110 determines that the recognition process is substantially impossible.

In a second example, processor 110 obtains recognition performance based on results of past recognition processes. For example, it is assumed that the position (absolute position) of the reference object is registered in advance in the map information 230 . Signals, signs, poles, white lines, guardrails, and the like are examples of reference objects. Object information 215 indicates the relative position of the reference object detected by the recognition process. By combining the object information 215 and the vehicle position information 240, the absolute position of the reference object detected by the recognition process is obtained. The processor 110 compares the reference objects registered in the map information 230 and the reference objects detected by the recognition processing to calculate the detection rate of the reference objects by the recognition processing. Processor 110 then determines the recognition performance based on the calculated detection rate. For example, if the detection rate is greater than or equal to the first threshold, processor 110 determines that the recognition performance is high. If the detection rate is less than the first threshold and greater than or equal to the second threshold, processor 110 determines that the recognition performance is moderate. If the detection rate is less than the second threshold and greater than or equal to the third threshold, processor 110 determines that the recognition performance is low. If the detection rate is less than the third threshold, processor 110 determines that the recognition process is substantially impossible.

In a third example, processor 110 obtains recognition performance based on the type of object recognized by the recognition process. For example, when at least a white line is recognized, the processor 110 determines that the recognition performance is "low" or higher. When an object in front of the automatic driving vehicle 1 is at least recognized, the processor 110 determines that the recognition performance is "medium" or higher. When objects in all directions around the autonomous vehicle 1 are recognized, the processor 110 determines that the recognition performance is high.

In a fourth example, processor 110 may obtain localization accuracy as recognition accuracy.

3-3. Route selection processing (step S330)
In step S320, processor 110 performs a "route selection process." In the route selection process, the processor 110 selects the target route RT from multiple route candidates according to the recognition performance of the recognition process. A plurality of route candidates are obtained from the route candidate information 300 described above. The recognition performance of recognition processing is obtained from the recognition performance information 400 described above.

As examples of multiple route candidates, consider a first route R1, a second route R2, and a third route R3 as shown in FIG. The first route R1 requires high recognition performance. The second route R2 requires less recognition performance than the first route R1. A third route R3 requires recognition performance equal to or lower than that of the second route R2.

For example, the first route R1 not only passes through a relatively wide arterial road, but also enters a relatively narrow community road. The third route R3 mainly passes through relatively wide arterial roads and does not enter relatively narrow community roads. The first route R1 is the longest and the third route R3 is the shortest. The second route R2 is intermediate between the first route R1 and the third route R3.

As another example, the first route R1 includes sections without a centerline. On the second route R2, at least pedestrians and vehicles are separated by white lines. Pedestrians and vehicles are separated by curbs and white lines on the third route R3.

FIG. 7 is a flowchart showing an example of route selection processing.

In step S332, processor 110 determines whether the recognition performance is high. If the recognition performance is high (step S332; Yes), the processor 110 selects the first route R1 as the target route RT (step S333). Otherwise (step S332; No), the process proceeds to step S334.

In step S334, processor 110 determines whether the recognition performance is moderate. If the recognition performance is moderate (step S334; Yes), the processor 110 selects the second route R2 as the target route RT (step S335). Otherwise (step S334; No), the process proceeds to step S336.

In step S336, processor 110 determines whether the recognition performance is low. If the recognition performance is low (step S336; Yes), the processor 110 selects the third route R3 as the target route RT (step S337). Otherwise (step S336; No), the process proceeds to step S338.

In step S338, the processor 110 performs abnormality handling. For example, processor 110 stops automatic driving vehicle 1 . Processor 110 may communicate with a remote assistance system and request remote assistance from a remote operator.

FIG. 8 is a flowchart showing another example of route selection processing. The first route R<b>1 requires recognition performance capable of recognizing objects in all directions around the autonomous vehicle 1 . The second route R2 requires at least recognition performance capable of recognizing an object in front of the autonomous vehicle 1 . The third route R3 requires at least recognition performance capable of recognizing white lines.

At step S331, the processor 110 determines whether it is raining or snowing based on the weather information 250. FIG. If it is not raining or snowing (step S331; No), the process proceeds to step S333. On the other hand, if it is raining or snowing (step S331; Yes), the process proceeds to step S332'.

At step S332', the processor 110 determines whether or not a peripheral object can be recognized. If the surrounding object can be recognized (step S332'; Yes), the processor 110 selects the first route R1 as the target route RT (step S333). Otherwise (step S332'; No), the process proceeds to step S334'.

At step S334', processor 110 determines whether or not a forward object can be recognized. If the forward object can be recognized (step S334'; Yes), the processor 110 selects the second route R2 as the target route RT (step S335). Otherwise (step S334'; No), the process proceeds to step S336'.

At step S336', the processor 110 determines whether the white lines are recognizable. If the white line is recognizable (step S336'; Yes), the processor 110 selects the third route R3 as the target route RT (step S337). Otherwise (step S336'; No), the process proceeds to step S338.

1 Autonomous Driving Vehicle 10 Autonomous Driving System 20 Sensor Group 21 Recognition Sensor 30 Travel Device 100 Control Device 110 Processor 120 Storage Device 200 Driving Environment Information 210 Surrounding Information 215 Object Information 220 Vehicle State Information 230 Map Information 240 Vehicle Position Information 250 Weather Information 300 route candidate information 400 recognition performance information RT target route

Claims (1)

An automated driving system applied to an automated driving vehicle that provides a mobility service that transports passengers,
comprising one or more processors,
The one or more processors are
Recognition processing for recognizing the surrounding situation of the automatic driving vehicle using a recognition sensor mounted on the automatic driving vehicle;
A route setting process for setting a route to a destination or a loop route as a target route for the automated driving vehicle,
The route setting process includes:
a route candidate acquisition process for acquiring a plurality of candidates for the target route;
a route selection process for selecting the target route from among the plurality of candidates according to the recognition performance of the recognition process;
including
In the route selection process, the one or more processors select candidates with shorter distances as the target route as the recognition performance declines. Automatic driving system.

Images